Google teilt seinen Mitarbeitern mit, dass die Kapazität alle sechs Monate verdoppelt werden muss, um die KI-Nachfrage zu decken

    https://arstechnica.com/ai/2025/11/google-tells-employees-it-must-double-capacity-every-6-months-to-meet-ai-demand/

    Share.

    28 Kommentare

    1. reminds me of that scene where Darth Vader tells the guy in charge the Emperor is coming….

      „we shall double our efforts…“

    2. The dumbest corporate bullshit. Let me guess they also reduced payroll by 8%.

    3. ARobertNotABob on

      Building data centers, capex, all doable.

      Cooling? That’s is HUGE amounts of water; impact to human’s supplies?
      Electricity? Similar. The current infrastructure is just about coping with current demand…and again, what impact will it have on existing local humans/community supplies, and costs?

    4. Meanwhile we can’t get VMs fast enough from them to meet our customer demand

    5. Good_Air_7192 on

      ‚also, please double your usage of AI every six months or you’re fired.“

    6. Anybody got a chessboard and some rice to show that that is a stupid idea?

    7. NebulousNitrate on

      There needs to be a harder push for efficiency. I work at a major cloud provider and a huge component should be on making the software infrastructure itself more efficient 

    8. neuronexmachina on

      Some more info from the original article: https://www.cnbc.com/2025/11/21/google-must-double-ai-serving-capacity-every-6-months-to-meet-demand.html

      >At an all-hands meeting on Nov. 6, Amin Vahdat, a vice president at Google Cloud, gave a presentation, viewed by CNBC, titled “AI Infrastructure,” which included a slide on “AI compute demand.” The slide said, “Now we must double every 6 months…. the next 1000x in 4-5 years.”

      > … Google’s “job is of course to build this infrastructure but it’s not to outspend the competition, necessarily,” Vahdat said. “We’re going to spend a lot,” he said, adding that the real goal is to provide infrastructure that is far “more reliable, more performant and more scalable than what’s available anywhere else.”

      >In addition to infrastructure build-outs, Vahdat said Google bolsters capacity with more efficient models and through its custom silicon. Last week, Google announced the public launch of its seventh generation Tensor Processing Unit called Ironwood, which the company says is nearly 30 times more power efficient than its first Cloud TPU from 2018.

      > … Google needs to “be able to deliver 1,000 times more capability, compute, storage networking for essentially the same cost and increasingly, the same power, the same energy level,” Vahdat said. “It won’t be easy but through collaboration and co-design, we’re going to get there.”

      I wasn’t previously familiar with Vahdat. Somewhat surprisingly, he seems like someone who actually knows what he’s talking about, so we’ll see how it goes: https://techsysinfra.google/aboutus/tsi-leaders/amin-vahdat/

    9. Hahaha….let’s see AI double capacity and *NOT* cause a public relations backlash as a result.

    10. If they double capacity every six months, all those Google colored pipes are going to burst. They won’t be able to scale at all.

      Maybe they should give their engineers Celeron powered laptops instead of MacBooks, and tell them to write AI code that can execute quickly on a Celeron. That seems like the better solution.

    11. All of big tech is going through this right now to varying degrees of absurdity. My team was hit hard and it is mind numbingly dumb how over-fitted AI is right now. The productivity gains must justify the cost of investment, or else.

    12. Healthy_Razzmatazz38 on

      if you want to get really hyped, they mention video generation, you’re not spending 100b a year on ai tiktoks, this is world models building synthetic training data for robotics.

    13. OkMemory9587 on

      Artificial demand, using an AI agent for every search demands computing power

    14. JustAtelephonePole on

      In humans, growth for the sake of growth is cancer. Sending flowers, bitches!

    15. GrolarBear69 on

      People don’t understand this is an arms race, and we absolutely will dump every watt of energy and compute into it before someone else does at all costs, or literally die.
      The winning AI will crush every other military, economy, and technology.
      There are legions of data scientists saying this is well beyond the atom bomb and far more lethal.

    16. Sure, that’s totally sustainable and not at all a top-heavy pillar prone to collapse AT ALL.

    17. snaggleboot on

      I can’t wait for the stain humanity leaves behind to be mostly ruins of datacenters

    Leave A Reply